Fix for fdo#39517. Looking through the emf+ records, it looks like the
assumption made that the minimum valid record length is 12 + non-0
length payload isnt' right, looks like that the minimum valid record
length is just 12, i.e. no need for a payload.

This needs some testing on other platforms!

About SAL_PROPERTIES in OOo Environment documentation: "If set
contains a bitfield activating various bug Xserver bug workarounds.
Not very useful nowadays."

Doing List cleanup, I ran into SaveSelection() and decided to delete it.
While I was in the class, I went ahead and cleanup them all up.

Is it really good Mac style programming practice to define your own
non-namespaced kThisAndThat constants that look like they might be
defined by some platform API?

And seriously, defining symbolic constants for small integers like
const kOneBit = 1 ?  These were not used, but if they had been would
it really be cleaner to use the identifier "kOneBit" instead of a
plain number 1?

Cases where small integers have magic meaning and don't stand for just
themselves are obviously different; there it makes sense to use
symbolic names and not the integer values. But in the case of things
like kOneBit that is not the case, as one might guesss from the name
it was supposed to be used to indicate a bit depth of one. Etc.

It wasn' used so no one should miss it.

Don't actually compile anything, though... Just the rscsfx.hxx is
needed.